Dataset Information

Big Data Analytical Approaches to the NACC Dataset: Aiding Preclinical Trial Enrichment.

ABSTRACT: BACKGROUND:Clinical trials increasingly aim to retard disease progression during presymptomatic phases of Mild Cognitive Impairment (MCI) and thus recruiting study participants at high risk for developing MCI is critical for cost-effective prevention trials. However, accurately identifying those who are destined to develop MCI is difficult. Collecting biomarkers is often expensive. METHODS:We used only noninvasive clinical variables collected in the National Alzheimer's Coordinating Center (NACC) Uniform Data Sets version 2.0 and applied machine learning techniques to build a low-cost and accurate Mild Cognitive Impairment (MCI) conversion prediction calculator. Cross-validation and bootstrap were used to select as few variables as possible accurately predicting MCI conversion within 4 years. RESULTS:A total of 31,872 unique subjects, 748 clinical variables, and additional 128 derived variables in NACC data sets were used. About 15 noninvasive clinical variables are identified for predicting MCI/aMCI/naMCI converters, respectively. Over 75% Receiver Operating Characteristic Area Under the Curves (ROC AUC) was achieved. By bootstrap we created a simple spreadsheet calculator which estimates the probability of developing MCI within 4 years with a 95% confidence interval. CONCLUSIONS:We achieved reasonably high prediction accuracy using only clinical variables. The approach used here could be useful for study enrichment in preclinical trials where enrolling participants at risk of cognitive decline is critical for proving study efficacy, and also for developing a shorter assessment battery.

SUBMITTER: Lin M

PROVIDER: S-EPMC5854492 | biostudies-literature | 2018 Jan-Mar

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Big Data Analytical Approaches to the NACC Dataset: Aiding Preclinical Trial Enrichment.

Lin Ming M Gong Pinghua P Yang Tao T Ye Jieping J Albin Roger L RL Dodge Hiroko H HH

Alzheimer disease and associated disorders 20180101 1

<h4>Background</h4>Clinical trials increasingly aim to retard disease progression during presymptomatic phases of Mild Cognitive Impairment (MCI) and thus recruiting study participants at high risk for developing MCI is critical for cost-effective prevention trials. However, accurately identifying those who are destined to develop MCI is difficult. Collecting biomarkers is often expensive.<h4>Methods</h4>We used only noninvasive clinical variables collected in the National Alzheimer's Coordinati ...[more]

PMID: 29227306

Similar Datasets

Project description:BACKGROUND:Longitudinal phenotypic data provides a rich potential resource for genetic studies which may allow for greater understanding of variants and their covariates over time. Herein, we review 3 longitudinal analytical approaches from the Genetic Analysis Workshop 19 (GAW19). These contributions investigated both genome-wide association (GWA) and whole genome sequence (WGS) data from odd numbered chromosomes on up to 4 time points for blood pressure-related phenotypes. The statistical models used included generalized estimating equations (GEEs), latent class growth modeling (LCGM), linear mixed-effect (LME), and variance components (VC). The goal of these analyses was to test statistical approaches that use repeat measurements to increase genetic signal for variant identification. RESULTS:Two analytical methods were applied to the GAW19: GWA using real phenotypic data, and one approach to WGS using 200 simulated replicates. The first GWA approach applied a GEE-based model to identify gene-based associations with 4 derived hypertension phenotypes. This GEE model identified 1 significant locus, GRM7, which passed multiple test corrections for 2 hypertension-derived traits. The second GWA approach employed the LME to estimate genetic associations with systolic blood pressure (SBP) change trajectories identified using LCGM. This LCGM method identified 5 SBP trajectories and association analyses identified a genome-wide significant locus, near ATOX1 (p = 1.0E(-8)). Finally, a third VC-based model using WGS and simulated SBP phenotypes that constrained the ? coefficient for a genetic variant across each time point was calculated and compared to an unconstrained approach. This constrained VC approach demonstrated increased power for WGS variants of moderate effect, but when larger genetic effects were present, averaging across time points was as effective. CONCLUSION:In this paper, we summarize 3 GAW19 contributions applying novel statistical methods and testing previously proposed techniques under alternative conditions for longitudinal genetic association. We conclude that these approaches when appropriately applied have the potential to: (a) increase statistical power; (b) decrease trait heterogeneity and standard error;

Dataset Information

Big Data Analytical Approaches to the NACC Dataset: Aiding Preclinical Trial Enrichment.

Publications

Big Data Analytical Approaches to the NACC Dataset: Aiding Preclinical Trial Enrichment.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets